Gemma 3 27B IT: VRAM by quant
27.4B params · KV field None · max context field 131072.
| Quant | Bits | Weight GB sketch |
|---|---|---|
| Q4_K_M | 4.85 bpw | ~17.0 GB |
| Q5_K_M | 5.7 bpw | ~19.7 GB |
| Q6_K | 6.6 bpw | ~22.7 GB |
| Q8_0 | 8.5 bpw | ~29.3 GB |
| MXFP4 | 4.25 bpw | ~15.1 GB |
| F16/BF16 | 16.0 bpw | ~54.8 GB |
Add KV for your context length in context VRAM. Overall fit: VRAM fit.
Educational. MoE active params may differ from total params.